Quantitation of Underivatized Omega-3 and Omega-6 Fatty ... - Dionex

Quantitation of Underivatized Omega-3 and Omega-6 Fatty Acids in Foods
by HPLC and Charged Aerosol Detection
Ian Acworth, Marc Plante, Bruce Bailey, and Christopher Crafts
Thermo Fisher Scientific, Chelmsford, MA, USA
Abstract
The omega fatty acids are a group of compounds that
include essential n-3 (omega-3, e.g., a-linolenic acid
[ALA]), n-6 (omega-6, e.g., linoleic and arachidonic acids),
and nonessential n-9, (omega-9, e.g., oleic
and erucic acids) analytes. The omega-3 fatty acids,
which also include eicosapentanoic acid (EPA) and
docosahexanoic acid (DHA), are required for normal
growth. Their consumption is purported to have a
number of health benefits: e.g., cancer prevention,
cardiovascular disease prevention, and improved immune
function. Although both omega-3 and -6 fatty acids can
give rise to eicosanoid-signaling molecules
(prostaglandins, prostacyclins, thromboxanes, and
leukotrienes), the omega-6 eicosanoids are generally
pro-inflammatory and may play a role in cardiovascular
disease, high blood pressure, and arthritis. It appears
that the amounts and balance of omega fatty acids in a
person’s diet affect their eicosanoid-controlled functions.
A proper balance of omega fatty acids in the diet
is important.
Traditionally, omega fatty acids are measured using
gas chromatography (GC). For foods, analytes are
extracted from the samples prior to hydrolysis to release
the fatty acids from their triglycerides, then converted
to their volatile methyl esters prior to analysis by GC.
This approach is tedious, time-consuming, and the
high temperatures can affect polyunsaturated fatty
acid stability. The Thermo Scientific Dionex Corona
charged aerosol detector provides a univeral massbased
approach that is sensitive, has a wide dynamic
range, and has a major advantage in that all nonvolative
analytes give similar response independent of
chemical structure. No derivatization is required, and
unlike UV detection, the analyte does not need to contain
a chromophore. Presented here is a simple and direct
high-performance liquid chromatography and charged
aerosol detection (HPLC-charged-aerosol-detector)
method that can be used to measure omega-3, -6 and -9
fatty acids in traditional and commercially produced meat,
fish, and oils, as well as over-the-counter supplements.
Introduction
The common determination of omega lipids in foods
comprises several steps, including extraction, hydroysis,
and derivitization for measurement by GC. GC works
well as a standard analytical tool for these determinations
but requires alteration of the sample, and it can also
adversely affect temperature-sensitive functional
groups on specialized lipids. HPLC with ultraviolet
detection requires use of low wavelengths, which limits
solvent selection and increases the likelihood of
matrix interference.
Shown here is a reversed-phase HPLC method for the
determination of omega fatty acids in oil/food samples
using a dual-gradient method and charged aerosol
detection. This combination enables determination of
many fatty acids in a single analysis, and without the
sample derivatization that is required for GC analysis.
Several fatty acids were analyzed, including six omega-3
fatty acids (stearidonic acid [SDA], eicosapentanoic
acid [EPA], a-linolenic acid [ALA], docosahexanoic acid
[DHA], docosapentanoic acid [DPA], eicosatrienoic acid
[ETA]), five omega-6 fatty acids (γ-linolenic acid [GLA],
arachidonic acid [Arach.], linoleic acid [LLA], adrenic
acid, and eicosadienioic acid [EDA]), and two omega-9
fatty acids (oleic and erucic acids). An omega-7 fatty
acid, 9E,14Z-conjugated linoleic acid (CLA), was also
included due to its cited importance as a nutrient. 1,2
Charged aerosol detection is mass sensitive and can be
added to HPLC or ultra HPLC (UHPLC) platforms. The
detector provides the most consistent response for all
nonvolatile and some semivolatile analytes of all HPLC
detection techniques. 3 It works by charging particles as
shown in Figure 1, and is not dependent on light
scattering, which has large variability and generally
lower sensitivity.
Charged aerosol detection has been successfully used
to characterize lipids of all classification, 4 including
phospholipids (reversed phase5 and normal phase6,7 ),
acylglycerides, phytosterols, free fatty acids, and free fatty
alcohols. This method complements the free fatty acids
method, using higher specificity for these analytes and a
Thermo Scientific Acclaim C30 reversed-phase column.

FIGURE 1. Schematic of the Corona charged aerosol
detector flow paths.
Method
Experimental Conditions
HPLC System: Thermo Scientific Dionex
UltiMate 3000 RSLC Dual Gradient
Mobile Phase A: Water/formic acid/mobile phase B
(900:3.6:100)
Mobile Phase B: Acetone/acetonitrile/tetrahydrofuran/
formic acid (675:225:100:4)
Column: Acclaim C30, 250 × 3 mm, 3 µm
Column Temp.: 30 °C
Flow Rate, Eluent
Gradient Pump: 1 mL/min
1
2
4
Flow Rate, Inverse
Gradient Pump: 1 mL/min
Eluent Gradient Pump Inverse Gradient Pump
Time (min) % B Time (min) % B
0.00 0.0 0.00 100.0
1.00 60.0 1.10 100.0
13.00 70.0 2.10 40.0
22.00 95.0 14.10 30.0
24.00 95.0 23.10 5.0
24.00 0.0 25.10 5.0
29.00 0.0 25.10 100.0
30.10 100.0
3
8
6
5
7
1. Liquid eluent enters from the HPLC system
2. Pneumatic nebulization occurs
3. Small droplets enter the drying tube
4. Large drops exit to drain
5. Dried particles enter the mixing chamber
6. Gas stream passes over corona needle
7. Charged gas collides with particles and transfers charge
8. High mobility species are removed
9. Remaining charged particles measured with a very sensitive electrometer
10. Signal is transferred to chromatographic data software
10
9
Injection Volume: 2.00 µL
Detector: Corona ultra RS charged
aerosol detector
Corona Filter: High
Corona Nebulizer Temp.: 15 °C
Total Run Time: 30.1 min
The system is configured so that the analytical pump
provides the eluent through the column, and the second
pump adds the solvents in a manner that is inverse of
the column gradient. In this manner, the composition of
the eluent entering the charged aerosol detector is
maintained at a constant percent organic to minimize
changes between relative response factors of the different
analytes. It was determined that the void time for the
analytical stream was 1.10 minutes greater than that for
the inverse-gradient stream, and this time was added
to the inverse-gradient pump program, as shown in the
table above. This configuration reduces the relative
response differences between analytes caused by the
changing percent organic in the column eluent.
Standard Preparation
All standards were dissolved and diluted to 2500 µg/mL
in alcohol, and diluted serially to a concentration of
6.25 µg/mL.
Sample Preparation
All solid fat samples (50–100 mg) were extracted in
1.2 mL methanol/chloroform (1:1) for 15 min using vortex
mixing. Extracts were then centrifuged to remove solids,
and 1.0 mL of extract was added to 4 mL of isopropanol/
water (3:2) and 1 mL of 5 M KOH.
All liquid oil samples (50 µL aliquot) were dissolved/
dispersed in 5 mL isopropanol/water (3:2) and 1 mL of
5 M KOH.
All samples were heated in an 80 °C water bath for 1 h
with occasional stirring. After samples were cooled, a
500 µL aliquot was removed and 25 µL of formic acid
was added to neutralize the sample.
Results and Discussion
A chromatogram of the 14 standards (2500 ng on
column [o.c.]) is shown in Figure 2.
2 Quantitation of Underivatized Omega-3 and Omega-6 Fatty Acids in Foods
26802

FIGURE 2. HPLC chromatogram of 14 omega-free
fatty acids.
50
pA
Peaks: 1. Stearadonic Acid 8. Docosapentanoic acid
2. Eicosapentanoic acid 9. 9E, 14Z- conjugated linoleic acid
3. α-Linolenic acid 10. Eicosatrienoic acid
4. γ-Linoleic acid 11. Adrenic acid
5. Docosahexanoic acid 12. Oleic acid
6. Arachidonic acid 13. Eicosadienoic acid
7. Linoleic acid 14. Erucic acid
1
2
3
4
5
8
6 7
-5
6.0 7.5 8.8 10.0 11.3 12.5 13.8 15.0 16.3 17.5 18.8 20.0 21.3 23.0
Peak retention times in min were found to be: SDA 11.8,
EPA 13.7, ALA 14.0, GLA 14.4, DHA 15.8, Arach. 16.4,
LLA 16.7, DPA 17.0, CLA 17.2, ETA 17.9, adrenic 18.9,
oleic 19.2, EDA 19.7, and erucic 23.1 minutes.
Use of the inverse gradient improved the consistency
of response across the analytes. The relative spread
of responses between the inverse-gradient method,
and the gradient-only method decreased by 42%,
calculated by ([inverse-gradient max–min] – [gradient-only
max–min]/[gradient-only max–min]). This indicates that the
use of the inverse gradient can provide a benefit to
quantitation when unknown peaks are present in a
sample chromatogram, providing for improved
quantitation estimates.
It was later discovered that a few of the omega fatty acid
standard materials had degraded during the course of
developing this method. This is the likely cause of the
wide range of responses seen in the calibration curves,
shown in Figures 3 and 4. In using a fresh standard for
DHA, the response was found to be similar to that of ALA.
The use of 10 mg/L of butylated hydroxyanisole (BHA)
in the standard and sample solutions may preserve the
dissolved analytes for a longer period of time without
affecting the chromatography.
Calibration curves, using the inverse gradient conditions
data, were fit using inverted second-order polynomials
resulting in correlation coefficients, r2 >0.9995. Triplicate
analyses provided peak area reproducibility with an RSD

Table 1. LOD and LOQ Values for
Omega-Free Fatty Acids by HPLC
Analyte LOD (ng o.c.) LOQ (ng o.c.)
SDA 9.7 32.5
EPA 11.5 38.5
ALA 13.2 43.9
GLA 13.4 44.6
DHA* 15.0 45.0
Arach. 21.4 71.4
LLA 10.4 34.7
DPA 8.4 28.1
CLA 10.6 35.2
ETA 5.5 18.2
Adrenic 10.1 33.8
Oleic 4.9 16.3
EDA 11.7 39.1
Erucic 7.9 26.3
* Many of the LOD and LOQ values found in Table 1 may actually be
lower than reported, due to standard degradation. Typical LOD values
for charged aerosol detection are 1–10 ng on column.
Several oils and fats were processed and analyzed.
In the initial oil hydrolyzation experiments, a solution
of ethanol/water (3:2) was used. It was found that the
oils did not hydrolyze well in this solution. With an
exchange of isopropanol for the ethanol, it was identified
that the isopropanol provided a greater yield of free fatty
acids, and this solution was used for the hydrolyzation of
the samples.
A chromatogram of hydrolyzed mustard oil is shown
in Figure 5. This oil sample was found to contain a
composition that is consistent with literature values8,9 shown in brackets: 51% erucic [41–50%], 11% oleic
[8–15%], 21% linoleic [13–20%], and 13%
a-linolenic acids.
FIGURE 5. HPLC chromatogram of hydrolyzed mustard oil
using a C30 150 × 4.5 mm, 5 µm column..
148
pA
Peaks: 1. α-Linolenic acid
5. Linoleic acid
6. Eicosatrienoic acid
8. Adrenic acid
9. Oleic acid
10. Eicosadienoic acid
15. Erucic acid
Other peaks are unidentified
1
23 4
5
6 7
0
12.9 14 15 16 17 18 19 20 21 22 23 24.1
Minutes
28111
8
9
10 11
12 13
14 16
4 Quantitation of Underivatized Omega-3 and Omega-6 Fatty Acids in Foods
15
A hydrolyzed fish oil-based, commercially available supplement
is shown in Figure 6. A large number of other
free fatty acids may also be present with 24 unidentified
peaks, in addition to the 14 evaluated in this study.
FIGURE 6. HPLC chromatogram of 20 µL hydrolyzed fish
oil with addition of 200 µL isopropanol to aid in solubility. A
total of 38 peaks were detected including all 14 standards.
Peaks:
4. Stearadonic Acid
8. Eicosapentanoic acid
9. α-Linolenic acid
10. γ-Linoleic acid
14. Docosahexanoic acid
100
15. Arachidonic acid
17. Linoleic acid
18. Docosapentanoic acid
19. 9E, 14Z- conjugated
linoleic acid
22. Eicosatrienoic acid
15. Adrenic acid
27. Oleic acid
30. Eicosadienoic acid
36. Erucic acid
pA
8
1 2 3
6
4
5 7
26
14
9
25 27
11
32
16 19
29
13
23
1718
30 33
10 12 15
20 22 28
31
35
21 24 34
37
36 38
-5
6.0 7.5 8.8 10.0 11.3 12.5 13.8 15.0 16.3 17.5 18.8 20.0 21.3 23.0
Minutes
28112-01
Twelve additional samples were hydrolyzed in a similar
manner and the results are shown in Table 2 and presented
based on the ratio of omega-3 to omega-6 ratio (highest
to lowest). Fish oil had the highest ratio, explaining its
use as an omega-3 oil supplement. Grass-fed beef was
determined to have a ratio of approximately 1, close
to literature values 0.3–0.7. 10 Interestingly, pasture-fed
chicken was determined to have a ratio of approximately
0.62, significantly different than the commercial chicken
omega fats ratio of approximately 0.05. 11

Table 2. Percent Compositions of HPLC
Analysis of Hydrolyzed Samples
Sample
Omega-3
(%)
Omega-6
(%)
Omega-9
(%)
3:6
Ratio
Fish Oil 92.0 3.2 4.5 29.0
Fish, Flax,
Borage
Supplement
81.0 13.0 5.9 6.1
Flax Oil 59.0 22.0 19.0 2.7
Beef, Grass-fed 23.0 24.0 52.0 0.95
Avocado Oil 15.0 22.0 64.0 0.68
Chicken, Pastured
25.0 40.0 35.0 0.62
Mustard Oil 14.0 23.0 63.0 0.62
Canola Oil 13.2 33.2 53.6 0.39
Olive Oil 4.8 13.0 82.0 0.37
Walnut Oil 12.0 75.0 13.0 0.16
Castor Oil 4.2 62.0 34.0 0.067
Safflower Oil 0.71 17.0 82.0 0.041
Sesame Oil 1.5 61.0 38.0 0.024
Corn Oil 1.6 72.0 26.0 0.022
Castor oil was tested to evaluate the method’s performance
with other oils. A chromatogram for hydrolyzed castor oil
is shown in Figure 7. Here, a single fatty acid, ricinoleate,
was found to have 83 area percent (uncalibrated) of
the fatty acids, which is similar to the reference value of
90%. 12 Ricinoleate, which contains a unique 12-hydroxy
group, elutes earlier than the other free fatty acids
quantified in this method.
FIGURE 7. HPLC chromatogram of hydrolyzed castor oil
with ricinoleate-free fatty acid at 8.545 min.
160
pA
1 2
-20
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
Minutes
28113
Conclusions
3
Peaks: 3. Rincinoleic acid
7. Linolenic acid
8. Linoleic acid
9. Adrenic acid
10. Oleic acid
11. Eicosadienoic acid
Other peaks are unidentified
4 5 6 7
8
9 10 11 1213 14
Using this method, it is possible to obtain quantitative
analyses of different omega-free fatty acids, including
omega-3, -6, -7, and -9. Samples were hydrolyzed to
separate the fatty acids from their glycerol backbone
and analyzed directly using HPLC with charged aerosol
detection. A wide variety of samples were analyzed,
including animal- and plant-based oils, over-the-counter
supplements, and meat fats. The mobile phase used
here is compatible with mass spectrometry, which allows
for the possibility of identifying unknown free fatty acids
that may exist in a sample.
5

References
1. Blankson, H.; Stakkestad, J.A.; Fagertun, H.; Thom,
E.; Wadstein, J.; Gudmundsen, O. J. of Nutr. 2000,
130, 2943–2948.
2. Kritchevsky D. Br. J. Nutr. 2000, 83 (5), 459–465.
3. Dependence of Response on Chemical Structure,
ESA—A Dionex Company, Application Brief
70-8913 Rev B.
4. Plante, M.; Bailey, B.; Acworth, I. The Use of
Charged Aerosol Detection with HPLC for the
Measurement of Lipids, Methods in Molecular
Biology, Donald Armstrong (ed.), 1, Volume 579,
Lipidomics, Part 2, 469–482.
5. Ramos, R.G.; Libong, D.; Rakotomanga, M;
Gaudin, K; Loiseau, P.M.; Chaminade, P.
J. Chromatogr A, 1209, 2008, 88–94.
6. Cascone, A.; Eerola, S.; Ritieni, A.; Rizzo, A.
J. Chromatogr A. 2006, 1120 (1-2) 211–220.
©2011 Thermo Fisher Scientific, Inc.
All trademarks are the property of Thermo Fisher Scientific Inc. and its subsidiaries.
This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.
Dionex Products: 1228 Titan Way, PO Box 3603, Sunnyvale, CA 94088-3603, (408) 737-0700
North America: U.S./Canada (847) 295-7500
South America: Brazil (55) 11 3731 5140
Europe: Austria (43) 616 51 25, Benelux (31) 20 683 9768 (32) 3 353 4294
Denmark (45) 36 36 90 90, France (33) 1 39 30 01 10, Germany (49) 61125 991 0
Ireland (353) 644 0064, Italy (39) 02 51 62 1267, Sweden (46) 8 473 3380,
Switzerland (41) 62 205 9966, United Kingdom (44) 1276 691722
Asia Pacific: Australia (61) 2 9420 5233, China (852) 2428 3282, India (91) 22 2764 2735,
Japan (81) 6885 1213, Korea (82) 2 2653 2580, Singapore (65) 6289 1190,
Taiwan (886) 2 875 6655
www.thermoscientific.com/dionex
LPN 2931-02 01/12
7. Sensitive, Single-Peak Phospholipid Quantitation
by NP-HPLC-CAD, ESA—A Dionex Company
Application Note #70-9094.
8. Banga, S.K. Breeding for Oil and Meal Quality.
In Chopra V.L., Prakash S. (Eds). Oilseed and
Vegetable Brassicas: Indian Perspective, 1996,
Chapter 11, 234–249.
9. Prakash, S.; Kumar, P.R.; Sethi, M.; Singh, C.;
Tandon, R.K. Mustard Oil: The Ultimate Edible Oil.
The Botanica 2000, 50, 94–101.
10. Daley, C.A.; Abbott, A.; Doyle, P.S.; Nader, G.A;
Larson, S. Nutrition Journal 2010, 9 (10), 1–12.
11. Based on data from USDA National Nutrient
Database for Standard Reference, Release 20,
http://jmyarlott.com/Food/health/Omega3/
Default.asp (last accessed 07 Jan 2011).
12. McKeon, T.A.; Lin J.T.; Stafford, A.E.; Adv. Exp. Med.
Biol. 1999, 464, 37–47.